Level control of Conical Tank Process using ANFIS based Model Reference Adaptive PID controller

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 03 Issue: 02 | Feb-2016 www.irjet.net p-ISSN: 2395-0072
© 2016, IRJET | Impact Factor value: 4.45 | ISO 9001:2008 Certified Journal | Page 387
Level control of Conical Tank Process using ANFIS based Model
Reference Adaptive PID controller
Sengeni D1, Abraham Lincon S2
1Research Scholar, Dept. of Electronics& Instrumentation, Annamalai University
2Professor, Dept. of Electronics& Instrumentation, Annamalai University, Tamilnadu, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - This paper describes the design and
implementation of ANFIS based Model Reference Adaptive
PID controller for a nonlinear Conical Tank Level System
(CTLS). The control structure is established on a CTLS. The
mathematical model of CTLS is developed and an ANFIS
based Model Reference Adaptive PID Controller is proposed
for this level system. The result of proposed controller is
compared with MRAC-PID and conventional PID to analyze
the performance in terms of integral square error and
Integral absolute error.Theresultsprovedthat the efficiency
of proposed controller.
Key Words: Conical tank, PID controller, ANFIS, MRAC.
1. INTRODUCTION
In most of the process industries, control of
chemical process system is challenging problems due to
their nonlinear dynamic behavior. In particular, one of the
non linear systems like conical tank is extensively used in
process industries, petrochemical industries, food process
industries and wastewater treatment industries. Conical
tank is highly non linear system because of cross section
with respect to level. Conventional Controllers are normally
used in process industries as they are simple and familiar to
the field operator but they give poorperformancebecauseof
tuning about one operating point. The variations in the
process parameters can be trounced by persistent tuning of
the controller parameters using adaptive intelligent
techniques like adaptive based control strategy.
One of the most frequently used adaptive control
technique is Model Reference Adaptive Control systems
(MRAC). This control system [1-4]hasreceivedconsiderable
attention, and many new approaches have been applied to
practical process. But conventional MRAC will not give
satisfactory response for non linear system because of its
adaptation procedure. So a soft computing technique is
introduced in the MRAC technique in order to overcome
these problems. A control strategythatenhancesa controller
with a self-learning capability for achieving prescribed
control objectives. Inthissense,anAdaptive-FuzzyInference
System (ANFIS) based MRAC PID architecture is employed
[5-6] so that a MRAC structure is builtforachievinga desired
input/output mapping. The learningmethodusedallowsthe
tuning of parameters both of the membership functions and
the consequents in a Sugeno–type inference system. In this
paper the Conical tank level system has beenconsideredasa
typical representative of inherentlynonlinearsystem,thusit
is an ideal choice for testing the modeling capability of the
ANFIS based MRAC algorithm.
The main contributions of this paper are the
performance of the ANFIS based MRAC control strategy on
the model of the conical tank level system through
simulation studies. In section 2 the process description of
conical tank is given. The MRAC is discussed in section3.The
design and structure of ANFIS control strategy is detailed in
section 4.Simulation results are analyzed in section 5.
Finally, section 6 is summing up of the entire work.
2. PROCESS DYNAMICS
Fig 1 shows the schematicConical tank level system.
Here Fi is the inlet flow rate to the tank, F0 be the outlet flow
rate from the tank, FL be the disturbance applied to the tank.

Fig -1: Conical Tank Level System
Fi - Inlet flow rate to the tank (m3 / min)
F0 - Outlet flow rate from the tank (m3 / min)
FL -Load applied to the tank (m3 / min)
H - Height of the conical tank (m)
h - Height of the liquid level in the tank at anytime
't' (m)
R - Top radius of the conical tank (m)
r - Radius of the conical vessel at a particular level
of height h(m)

A -Area of the conical tank (m2)
The nominal operating level h is given by
(1)
= (2)
At any level (h) = (3)
Equating (2) and (3)
r = (4)
Cross sectional area of the tank at any level(h)is
(5)
Substitute (4) in (5)
(6)
Also (7)
Substituting (7) in (1)
(8)
(9)
From equation (8)
(10)
Where
=Nominal value of outflow rate
Hence the transfer function of the above system is
= (11)
where,
h and U are nominal values of PV and MV from
equation (11)
Time constant of the level process (12)
The gain constant of the level process k =
3. MODEL REFERENCE ADAPTIVE PID CONTROL
When the plant parameters and the disturbanceare
varying slowly or slower than the dynamic behavior of the
plant, then a MRAC control can be used.Themodel reference
adaptive control scheme is shown in figure 2. The
adjustment mechanism uses the adjustable parameter
known as control parameter θ to adjust the controller
parameters. The tracking error and the adaptation law for
the controller parameters were determined by MIT rule.
Reference
Model
Parameter Adjustment
Mechanism
Controller
Conical Tank
System
uc
Ym
Yp
ANFIS
Fig- 2: Structure of Model Reference Adaptive Controller
MIT (Massachusetts Institute of Technology) Rule isthatthe
time rate of change of θ is proportional to negative gradient
of the cost function (J), that is:
The adaptationerror .Thecomponentsof
are the sensitivity derivatives of theerrorwith respectto the
adjustable parameter vector θ. The parameter γ is known as
the adaptation gain. The MIT rule is a gradient scheme that
aims to minimize the squared model error from cost
function.
Model reference Adaptive Control: The goal of this section
is to develop parameter adaptation laws for a PID control
algorithm using MIT rule.
The reference model for the MRAC generates the desired
trajectory , which the level has to follow. Standard
second order differential equation was chosen as the
reference model given by
Consider also the adaptive law of MRAC structure taken as
the following form
(16)
Where; , is proportional gain, is integral
gain, is derivative gain and is a unit step input. In the
laplace domain, equation (18) can be transformed to
It is possible to show that applying this control law to the
system gives the following closed loop transfer function:

Apply MIT gradient rules for determining the value of PID
controller parameters . The tracking error
equation satisfies:
The exact formulas that are derived using the MIT rule
cannot be used. Instead some approximations are
required. An approximation made which valid when
parameters are closed to ideal value is as follows:
Denominator of plant, Denominator model reference then,
from gradient method.
Where:
Then the approximate parameter adaptation laws are as
follows
Above equations show the change in PID controller
parameters with respect to time. By assuming the reference
model has 5% maximum overshoot, settling time of 30
seconds and rise time of about 1 seconds, the second order
transfer function of the Model Reference as follows
4. ANFIS MODEL REFERENCE ADAPTIVEPID CONTROL
ANFIS stands for Adaptive Neural Fuzzy Inference
System. Using a given input/output data set, ANFIS
constructs a Fuzzy Inference System (FIS) whose
membership function parameters are adjusted using back
propagation algorithm in combination with a least squares
technique. This allows fuzzy system to learn from the data.
The Takagi-Sugino ANFIS architecture is shown in Figure 3.
The circular nodes represent nodes that are fixed whereas
the square nodes are nodes that have parameters to be
learnt.
Fig- 3: ANFIS Architecture for Takagi-sugeno system
ANFIS has rules of the form:
If x is A1 and y is B1 THEN f1 = p1x + q1y + r1
If x is A2 and y is B2 THEN f2 = p2x + q2y + r2
FIS is trained using combination of least squares and back
propagation. The entire sugeno system consists of five
layers and the relationship between the i/o of each layer
is summarized in [11]. ANFIS based MRAC PID control of
CTLS is shown in Figure 4.
Reference
Model
Parameter Adjustment
Mechanism
Controller
Conical Tank
System
uc
Ym
Yp
ANFIS
Fig- 4: ANFIS based MRAC PID control of CTLS.
5.RESULTS AND DISCUSSION
Performances of proposedcontrollerareanalyzedusing step
input at various level in the CTLS.. Initially the tank is
maintained at 30 % operating level, after that, a step size of
20 % of level is applied to control loop with ANFIS based
MRAC-PID control strategy. In the same way, test runs of
MRAC-PID and conventional PID control values are carried
out and their responses are presented inFigure.5. Itisfound
that in ANFIS based MRAC-PID makes the system to settle
with minimum overshoot at all.
To validate the performance proposed controller,
the same procedure is repeated for 70% and 90% level and
given in the same figure. From the results, the performances
are analyzed in terms of ISE and IAE are tabulatedinTable 1.
From Fig 5, it is also clear that proposedcontrollertracksthe
set point quickly without any oscillations. The magnified
view of output response is presented in Figure 6.Theresults
prove that ANFIS based MRAC-PID controller is appropriate

for non linear process as it has least error values than the
other controller strategies.
0 100 200 300 400 500 600 700 800
20
30
40
50
60
70
80
90
100
Time (sec)
Level(%)
Operating point
PID
MRAC-PID
ANFIS based MRAC-PID
Fig-5: Comparison of performances of PID, MRAC-PID
and ANFIS based MRAC-PID for conical tank level
process.
400 410 420 430 440 450 460 470 480 490 500
50
55
60
65
70
75
80
Time (sec)
Level(%)
Operating point
PID
MRAC-PID
ANFIS based MRAC-PID
Fig-6: Magnified view of output response.
Table.1. Performance Indices at different Operating
range.
Controller
ISE
OP
(3050%)
OP
(50to70%)
OP
(70to90%)
PID 1171 2456 4290
MRAC-PID 552.7 866.7 1083
ANFIS
MRAC-PID
497.7 737.6 964.2
IAE
Controller
OP
(30to50%)
OP
(50to70%)
OP
(70to90%)
PID 166.9 354.1 594.9
MRAC-PID 59.81 102.2 128.1
ANFIS
MRAC-PID
53.15 85.38 114.3
6.CONCLUSION
In this paper, ANFIS based MRAC-PID control
strategy has been developed and implemented for a conical
tank level system. This methodissuitablefor processcontrol
applications with a large delay, where a conventional PID
controller yield a poor performance. The simulation results
are furnished to illustrate the efficiency of proposed
controller with those of MRAC-PID and PID control
approaches. The performance indices are also proved that
the proposed controller gives a superior performance than
the existing control strategies.
REFERENCES
1. Hans Butler, Ger Honderd, and Job van Amerongen,
“Model ReferenceAdaptiveControl ofa Direct-Drive
DC Motor” IEEE Control Systems Magazine, 1989,pp
80-84.
2. S. Vichai, S. Hirai, M. Komura and S. Kuroki, “Hybrid
Control-based Model Reference Adaptive Control”
lektronika Ir Elektrotechnika”, Nr. Vol.3(59), 2005,
pp 5-8.
3. Chandkishor, R., Gupta, O, “Speed control of DC
drives using MRAC technique”, 2nd International
Conference on Mechanical and Electrical
Technology, ICMET, Sep 2010, pp 135 – 139.
4. Missula Jagath Vallabhai, Pankaj Swarnkar, D.M.
Deshpande, “Comparative Analysis Of PI Control
and Model ReferenceAdaptiveControl basedVector
Control Strategy For Induction Motor Drive,”IJERA,
Vol. 2(3), Jun 2012, pp 2059-2070.
5. K. Premkumar and B.V. Manikandan, “Fuzzy PID
supervised online ANFIS based speed controller for
brushless DC motor,” Neuro computing,Vol.157(1),
June 2015, pp.76–90
6. Hidayat, Pramonohadi,S. Sarjiya andSuharyanto,“A
comparative study of PID, ANFIS and hybrid PID-
ANFIS controllers for speed control of BrushlessDC
Motor drive,” IEEE proceedings on International
Conference on Computer, Control, Informatics and
Its Applications, IC3INA, 2013,Jakarta,, pp.117 –
122,.

Level control of Conical Tank Process using ANFIS based Model Reference Adaptive PID controller

More Related Content

What's hot (20)

Similar to Level control of Conical Tank Process using ANFIS based Model Reference Adaptive PID controller (20)

More from IRJET Journal (20)

Recently uploaded (20)

Level control of Conical Tank Process using ANFIS based Model Reference Adaptive PID controller